Case Number : Case 3011 - 20 January 2022 Posted By: Saleem Taibjee

Spitzoid nevi with NTRK fusion are usually compound, with rosettation or schwannomatous pattern. Here I see some rosettation in junctional nests. And the spitzoid features aren’t so forthright. But why not? It could be a NTRK fused ( NTRK1? or NTRK3? and partner? MYO5A? or ETV6? ) Spitz nevus. MYO5A are usually schwannomatous.  ETV6 are more epithelioid and atypical. 

Young age, symmetrical, yes some pagetoid spread but otherwise bland, maybe a Kamino body in there. Benign nevus with mild Spitz atypia, completely excised.

Appears symmetrical. There is bridging and some pagetoid scatter. May be a dysplastic nevus. 

Hi. Yes, I too favoured a benign Spitz, with some Reed-like features. There was quite prominent Pagetoid spread which as we know is 'allowable' in Spitz; the lesion is quite small with good lateral demarcation and no other overtly worrying features. The background melanin pigment made immunohistochemistry difficult to interpret, but I did think there is genuine staining for NTRK (see below). This particular stain is known to show weak, but genuine positive staining, corresponding to underlying NTRK fusions. Importantly, BRAFV600E IHC was negative, making an early superficial spreading melanoma even less likely. I also include ALK and ROS-1 in my Spitz panel, both negative in this case.

A recent study showed most pigmented spindle cell naevi/Reed (78%) are fusion-driven, commonly involving NTRK3 (57%). Conversely, BRAF V600E mutations are not characteristic of PSCN but are frequent in melanoma: Genomic fusions in pigmented spindle cell nevus of Reed. Am J Surg Pathol. 2018;42:1042–1051 https://pubmed.ncbi.nlm.nih.gov/29794873/.

There is a follow-on study: Retrospective Cohort: Genomic Differences Between Pigmented Spindle Cell Nevi of Reed and Reed-Like Melanomas https://pubmed.ncbi.nlm.nih.gov/32000215/ in which 3 of the PSCN lacking a fusion did show BRAF mutations, but 2 were 'non-canonical' mutations, and only one was BRAFV600E.

I am very interested to hear about Vincenzo's observation of rosette architecture of the junctional nests which I must admit I had not picked up. It is very interesting to learn about how we can directly correlate morphological clues on H&E with associated immunohistochemistry and molecular pathways.

I learned, from Arnaud de la Fouchardiere, many morphological features correlate with genomic mutations. NTRK3 mutated spitz neoplasms show often schwannomatouse features, because the partner of NTRK3, MYO5A, forces the protein produced by the fusion to remain in the cytoplasm membrane, and mainly in dendritic processes, so the cells are slender, neurinoma-like. but paraganglioma-like, also. Another partner of NTRK3, ETV6, forces the protein to go and remain into the nucleus, so the cells are swollen and epithelioid. However I’m not an expert in these things, so I’ve to search the web for useful literature in this regard. This things fascinate me but they are a bit far from my usual practical approach to diagnosis. 

Thanks Saleem for your educative comment.